Patient Adaptive Model for Medical Image Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical imaging technologies face challenges in accurately registering and fusing images from different modalities, such as ultrasound and MRI, due to patient movement, which affects the precision and accuracy of diagnostic and procedural images.
Innovation Solution
A medical imaging apparatus and method that utilizes a patient adaptive object model to correct for patient movement by transforming images based on differences in object shape and applying boundary conditions, incorporating a statistical object model and physical model to align and fuse images from different modalities, thereby improving image registration accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image registration is performed between different modalities without correcting for patient movement, then the registration process is simple and fast, but the registration accuracy and precision deteriorate due to patient movement
Solution Approach 1:
A patient adaptive object model is introduced as an intermediary between the actual patient anatomy and the image registration process. This model incorporates statistical object models and physical models to represent and correct for patient movement, serving as a mediator that translates between different imaging modalities while compensating for movement artifacts
Solution Approach 2:
The system performs preliminary actions by generating a patient adaptive object model before image registration occurs. This model is created by applying statistical object models and physical models to predict and correct patient movement, allowing the registration process to work with pre-corrected data rather than raw movement-affected images
2Measurement precision
If a statistical object model is applied to create a patient adaptive object model, then the correction for patient movement is more accurate, but the processing time and computational complexity increase
Solution Approach 1:
Statistical object models are pre-computed and stored before the actual imaging and registration process. These pre-computed models capture typical patient movement patterns and can be quickly applied during registration without requiring real-time computation, thus maintaining high accuracy while reducing processing time
Solution Approach 2:
The system changes parameters by using pre-determined statistical models that capture movement patterns through key parameters rather than full computational simulations. This parameter-based approach maintains correction accuracy while significantly reducing the computational burden and processing time
Data Source
AI summary
The medical imaging apparatus includes a storage unit configured to store a patient adaptive object model in which a statistical object model related to transformation of an object caused by a periodic movement is applied to characteristics of the object of a target patient; an image obtainer configured to acquire a first image of the target patient; and an image fuser configured to transform a second image of the target patient based on differences between a shape of the object extracted from the patient adaptive object model and a shape of the object displayed on the first image.